Lower plasma trans-4-hydroxyproline and methionine sulfoxide levels are associated with insulin dysregulation in horses

Abstract

Background: Insulin dysregulation in horses is a metabolic condition defined by high insulin concentrations in the blood and peripheral insulin resistance. This hyperinsulinemia is often associated with severe damage in the hooves, resulting in laminitis. However, we currently lack detailed information regarding the potential involvement of particular metabolic pathways in pathophysiological causes and consequences of equine insulin dysregulation. This study aimed to assess the dynamic metabolic responses given to an oral glucose test (OGT) in insulin-sensitive and insulin-dysregulated horses by a targeted metabolomics approach to identify novel metabolites associated with insulin dysregulation.

Results: Oral glucose testing triggered alterations in serum insulin (26.28 ± 4.20 vs. 422.84 ± 88.86 μIU/mL, p < 0.001) and plasma glucose concentrations (5.00 ± 0.08 vs. 9.43 ± 0.44 mmol/L, p < 0.001) comparing basal and stimulated conditions after 180 min. Metabolome analyses indicated OGT-induced changes in short-chain acylcarnitines (6.00 ± 0.53 vs. 3.99 ± 0.23 μmol/L, p < 0.001), long-chain acylcarnitines (0.13 ± 0.004 vs. 0.11 ± 0.002 μmol/L, p < 0.001) and amino acids (2.18 ± 0.11 vs. 1.87 ± 0.08 μmol/L, p < 0.05). Kynurenine concentrations increased (2.88 ± 0.18 vs. 3.50 ± 0.19 μmol/L, p < 0.01), whereas spermidine concentrations decreased during OGT (0.09 ± 0.004 vs. 0.08 ± 0.002 μmol/L, p < 0.01), indicating proinflammatory conditions after oral glucose load. Insulin dysregulation was associated with lower concentrations of trans-4-hydroxyproline (4.41 ± 0.29 vs. 6.37 ± 0.71 μmol/L, p < 0.05) and methionine sulfoxide (0.40 ± 0.06 vs. 0.87 ± 0.13 μmol/L, p < 0.01; mean ± SEM in insulin-dysregulated vs. insulin-sensitive basal samples, respectively), two metabolites which are related to antioxidant defense mechanisms.

Conclusion: Oral glucose application during OGT resulted in profound metabolic and proinflammatory changes in horses. Furthermore, insulin dysregulation was predicted in basal samples (without OGT) by pathways associated with trans-4-hydroxyproline and methionine sulfoxide, suggesting that oxidative stress and oxidant-antioxidant disequilibrium are contributing factors to insulin dysregulation. The present findings provide new hypotheses for future research to better understand the underlying pathophysiology of insulin dysregulation in horses.

Keywords: Horses; Insulin dysregulation; Insulin sensitivity; Metabolome; Metabolomics; Oral glucose test.

Fig. 1

Increase of (a) serum insulin (repeated measure two-way ANOVA (rmTWA) factor oral glucose testing (OGT) p < 0.001) and of (b) glucose (rmTWA factor OGT p < 0.001) concentration in insulin-sensitive (IS) and insulin-dysregulated (ID) horses during OGT. Measurements were carried out at three levels of OGT: immediately before glucose administration (BASAL), and 120 min (OGT-120) and 180 min (OGT-180) after administration. Means ± SEM are shown, n = 10, significant differences between IS and ID are indicated as ***(p < 0.001), and a trend between IS and ID is indicated as ⁺(0.05 < p < 0.1). a, b = different superscripts indicate significant differences between basal and challenged conditions as the main OGT effect in both ID and IS horses (Bonferroni post hoc test)

Fig. 2

a Principal component analysis scores plot showing metabolic profiles of insulin-sensitive (IS) and insulin-dysregulated (ID) horses sampled before glucose administration (BASAL), and 120 min (OGT-120) and 180 min (OGT-180) after administration. b Number of significant metabolites detected by repeated measure two-way ANOVA with false discovery rate correction, analyzing the effect of OGT, insulin sensitivity status (IS vs. ID) and respective interactions. c Heatmap illustrating relative concentrations of top 22 significant metabolites during OGT: BASAL, OGT-120 and OGT-180 for IS and ID horses

Fig. 3

Metabolites of importance which were significantly affected by OGT in insulin-sensitive (IS) and insulin-dysregulated (ID) horses. Concentration of (a) hexoses (repeated measure two-way ANOVA (rmTWA), factor OGT p < 0.001) and (b) kynurenine (rmTWA, factor OGT p < 0.01) increased, while (c) spermidine (rmTWA, factor OGT p < 0.01), (d) sum of all proteinogenic amino acids (rmTWA, factor OGT p < 0.05) (e) sum of short-chain (C2, C3, C4 and C5) acylcarnitines (rmTWA, factor OGT p < 0.001) and (f) sum of long-chain (C16, C16:1, C16:2, C:18, C18:1, C:18:2) acylcarnitines (rmTWA, factor OGT p < 0.001) decreased. Insulin status had no significant effect. Means ± SEM are shown, n = 10; a, b, c = different superscripts indicate significant differences between basal and challenged conditions as the main OGT effect in both ID and IS horses (Bonferroni post hoc test)

Fig. 4

a Volcano plot highlighting metabolites of importance (trans-4-hydroxyproline p = 0.0036; methinonine sulfoxide p = 0.0134) which were different between insulin-sensitive (IS) and insulin-dysregulated (ID) horses under unchallenged conditions (only BASAL) (b) Trans-4-hydroxyproline (repeated measure two-way ANOVA (rmTWA), factor insulin sensitivity status p < 0.05, factor OGT p = 0.06) and (c) methionine sulfoxide (rmTWA, factor insulin sensitivity status p < 0.05, factor OGT p = 0.06) concentrations shown in serum of IS and ID horses at three levels of OGT: before glucose administration (BASAL), and 120 min (OGT-120) and 180 min (OGT-180) after administration (means ± SEM, n = 10). Significant differences between IS and ID are indicated as ** (p < 0.01), * (p < 0.05). a, b = different superscripts indicate significant differences (p < 0.05) between basal and challenged conditions in IS horses, while no OGT effect was observed (as indicated by A) in ID horses (Bonferroni post hoc test)